Diphenylhexatrienes as photoprotective agents for ultrasensitive fluorescence detection

Article abstract of DOI:10.1021/jp909033x

Given the particular importance of dye photostability for single-molecule investigations, fluorescence fluctuation spectroscopy, and laser-scanning microscopy, refined strategies were explored for enhancing the fluorescence signal by selectively quenching

Full text of DOI:10.1021/jp909033x

J. Phys. Chem. A 2010, 114, 4099–4108
4099
Diphenylhexatrienes as Photoprotective Agents for Ultrasensitive Fluorescence Detection
Daniela Pfiffi,|,† Brigitte A. Bier,|,‡ Christel M. Marian,*,§ Klaus Schaper,*,‡ and
Claus A. M. Seidel*,†
Chair for Molecular Physical Chemistry, Heinrich-Heine-UniVersita¨t Du¨sseldorf, UniVersita¨tsstrasse 1,
D-40225 Du¨sseldorf, Germany, Group for Organic Photochemistry, DiVision of Organic Chemistry, Institute
for Organic Chemistry and Macromolecular Chemistry, Heinrich-Heine-UniVersita¨t Du¨sseldorf,
UniVersita¨tsstrasse 1, D-40225 Du¨sseldorf, Germany, and Institute of Theoretical and Computational
Chemistry, Heinrich-Heine-UniVersita¨t Du¨sseldorf, UniVersita¨tsstrasse 1, D-40225 Du¨sseldorf, Germany
ReceiVed: September 18, 2009; ReVised Manuscript ReceiVed: February 1, 2010
Given the particular importance of dye photostability for single-molecule investigations, fluorescence
fluctuation spectroscopy, and laser-scanning microscopy, refined strategies were explored for enhancing
the fluorescence signal by selectively quenching the first excited triplet state of the laser dye Rhodamine
123 (Rh123). The strategy is to quench the T1 state by Dexter triplet energy transfer, while undesired
quenching of the singlet state via Fo¨rster or Dexter singlet energy transfer and the generation of free
radicals through electron transfer should be avoided. Diphenylhexatrienes (DPHs) were tested in ethanol
for their beneficial effects as a novel class of photoprotective agents using fluorescence correlation
spectroscopy. A library of DPHs with electron-donating (dimethlyamino) and withdrawing substituents
(e.g., trifluormethyl) was synthesized to optimize the electronic properties. Quantum chemical calculations,
optical spectroscopy, and cyclic voltammetry were used to determine the electronic properties. The
computed T1 emission energy of Rh123 and the T1 excitation energies of all DPHs allow for exergonic
triplet energy transfer to the quencher. The parent compound quenches the T1 state of Rh123 nearly
diffusion controlled (4.9 × 109 M-1 s-1). All electron-deficient DPHs significantly increase (3×) the
fluorescence rate of Rh123 by reducing the triplet state population and by avoiding the formation of
other long-lived dark radical states. The quenching constants are reduced by more than a factor of 2, if
substituents with increasing size or electronegativity are introduced. The beneficial effect of triplet
quenching of substituted DPHs is governed by a delicate interplay of steric, electronic, and intermolecular
Coulombic effects.
1. Introduction
Here, we suggest using quenchers Q, which selectively
quench the triplet state, for this purpose. In principle, three
different quenching mechanisms are feasible, quenching by
dipole-dipole interaction (Fo¨rster mechanism),21-23 quenching
by electron exchange interaction (Dexter mechanism),24 and
quenching by redox processes.25-28 Among these, the Fo¨rster
mechanism can be excluded for triplet energy transfer, since
its efficiency depends on the oscillator strength of the involved
transitions.
Due to their high fluorescence quantum yields and high
photostability, rhodamines1-3 are important tools in confocal
and superresolution microscopy,4,5 single-molecule spectrosco-
py,6 and fluorescence correlation spectroscopy (FCS).7 Their
beneficial properties are based on small singlet-triplet inter-
system crossing (ISC) rates (kISC ≈ 105-106 s-1). Despite the
low probability of this process compared to fluorescence
emission (k0 ≈ 109-1010 s-1), ISC to the long-lived first excited
triplet state (3F1) is a major limitation in single-molecule
spectroscopy where high repetition rates of excitation and
fluorescence are required. Under these conditions, the fluores-
cence dyes F will accumulate in the triplet state at higher
irradiances and thus the signal saturates. Furthermore, the triplet
state is involved in the photochemical destruction of the
chromophore. Many attempts have been made to reduce the
triplet lifetime in order to increase signal strength and photo-
stability of free dyes8,9 and bichromophores.10-13 To this end,
various cocktails and procedures have been used.8,14-20
In order to achieve efficient triplet excitation energy
transfer (TEET) by the Dexter mechanism, certain conditions
have to be fulfilled (Figure 1A). The quenching should be
exergonic, and, at the same time, a significant overlap of the
phosphorescence spectrum of the donor fluorophore (3F1f1F0)
and the triplet-absorption spectrum of the quencher (3Q1r1Q0)
is necessary.29,30 Furthermore, to avoid singlet excitation
energy transfer (SEET) by the Fo¨rster and Dexter mecha-
nisms, the first excited singlet state of the quencher (1Q1)
should be located energetically above the fluorescing state
(1F1) of the fluorophore (Figure 1B). Additionally, undesired
singlet quenching by oxidation (ox) or reduction (red) of the
fluorophore has to be prevented (Figure 1C). The thermo-
dynamic properties of the latter processes can be estimated
using the Rehm-Weller equation25,26 based on absorption
spectra and ground-state redox properties:
* To whom correspondence should be addressed. E-mail: cm@
theochem.uni-duesseldorf.de; fax: +49 211 81 13466; tel: +49 211 81 13209
(C.M.M.). E-mail: schaper@klaus-schaper.de; fax: +49 211 81 14324; tel:
+49 211 81 12571 (K.S.). E-mail: cseidel@gwdg.de; fax: +49 211 81
12803; tel: +49 211 81 15881 (C.A.M.S.).
† Chair for Molecular Physical Chemistry.
‡ Institute for Organic Chemistry and Macromolecular Chemistry.
§ Institute of Theoretical and Computational Chemistry.
| These authors contributed equally to this work.
10.1021/jp909033x  2010 American Chemical Society
Published on Web 03/10/2010

4100 J. Phys. Chem. A, Vol. 114, No. 12, 2010
∆GET ) Eox - Ered - ∆En0 - ∆G0(ε)
Pfiffi et al.
tion energy of radical ions in a complex and (II) the effects
of ion solvation. Because the investigated electron transfer
reactions start with Rhodamine 123 (Rh123) in its cationic
ground state and end up with a cationic quencher molecule,
we neglect ∆G0(ꢀ) describing the solvation and the Coulomb
interactions of the solvent-separated ion pair. Moreover, it
is likely in protic solvents that proton-coupled reactions occur,
which increase the feasibility of quenching.31 The driving
force for triplet quenching by redox processes is always
smaller than the one for the singlet case, because the
excitation energy ∆En0 is smaller for the triplet case (eq 1,
Figure 1, and Tables 1 and 3). Thus selective triplet
quenching by this mechanism is impossible.
(1)
where Eox and Ered are the oxidation and reduction potentials
of the electron donor and acceptor, respectively, ∆En0 is the
singlet or triplet excitation energy. ∆G0(ꢀ) contains two
correction terms based on the Born equation: (I) the interac-
Carotenoids with 11-13 conjugated double bonds are present
in photosynthetic centers.32 Besides functioning as antenna
complexes, they are known to quench the T1 state of chloro-
phylls selectively, which otherwise would be an efficient
sensitizer for singlet oxygen. However, for our purpose of
quenching the triplet population of fluorescence dyes, these long-
chained carotenoids are not suited because their optically bright
11B+u state (S2 or S3) is low-lying. In addition, a dipole-forbidden
transition to the S1 state (21Ag-r11Ag-) takes place in the low-
energy regime.33-35 This dark 21A-g state might be involved in
undesired singlet quenching. Possible candidates with the desired
quenching properties are carotenoids with shorter conjugation
lengths. Here, aromatic pseudocarotenoids were investigated
because these systems allow us to easily adjust the electronic
properties by changing the substituents of the aromatic system.
For this purpose, a small library of diphenylhexatrienes
(DPHs) with different electronic properties was synthesized
(Figure 2A). Electron-donating and electron-withdrawing groups
were introduced into the aromatic system in order to create
compounds that were either (I) electron-rich [(all-E)-1,6-bis-
[4-(dimethylamino)phenyl]hexa-1,3,5-triene (Me2N-DPH)],37,38
(II) neutral [(all-E)-1,6-diphenylhexa-1,3,5-triene (DPH)37,39,40
or(all-E)-1,6-bis-(4-fluorophenyl)hexa-1,3,5-triene(F-DPH)],39,41-44
or (III) electron deficient [(all-E)-1,6-bis-[4-(trifluorometh-
yl)phenyl]hexa-1,3,5-triene (CF3-DPH),45 (all-E)-1,6-bis-(4-
cyanophenyl)-hexa-1,3,5-triene (CN-DPH),37 or (all-E)-1,6-bis-
[3,5-bis(trifluoromethyl)phenyl]hexa-1,3,5-triene
(bis-CF3-
DPH)]. The electronic structures of these DPHs were determined
using theoretical methods, the photophysical and electrochemical
properties of the compounds were investigated, and their ability
to selectively quench the triplet state was measured.
2. Experimental Procedures and Computational Details
2.1. Compounds. The details of the synthesis and the
analytical characterization of all products are given in the
Supporting Information (SI, Section 1). In order to synthesize
the desired compounds, we chose a strategy starting from
substituted benzaldehydes and (E)-tetraethyl but-2-ene-1,4-
diyldiphosphonate.40 As a typical example, we describe the
synthesis of bis-CF3-DPH. (E)-Tetraethyl but-2-ene-1,4-diyl-
diphosphonate (1.62 g, 4.93 mmol) was dissolved in 40 mL of
dry tetrahydrofuran (THF). A 60% dispersion of sodium hydride
in mineral oil (0.39 g, 9.8 mmol) was added with 10 mL of dry
THF, and the mixture was stirred at room temperature for
30 min. To this mixture a solution of 2.73 g (11.3 mmol) of
3,5-bis(trifluoromethyl)benzaldehyde in 40 mL of dry THF was
added dropwise over 5 min, and the mixture was stirred for
another 30 min. A second dose of a 60% dispersion of sodium
hydride in mineral oil (0.37 g, 9.3 mmol) was added with 10
mL of dry THF, and the mixture was stirred for 3 d at room
temperature. Hydrolysis with 75 mL water causes the crude
Figure 1. Overview of the energy of the states of fluorophore F and
quencher Q involved in quenching processes. Depending on the nature
of the quencher, these energies may vary (gray boxes). The desired
values for these energies are marked by a black bar. (A) Energy balance
for SEET and TEET. (B) Energies of individual states. (C) Possible
electron transfer pathways.

DPH as Photoprotective Agent for Fluorescence Detection
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4101
Figure 3. Absorption (red) and fluorescence (blue) spectra in THF
(circles) and multi-Gauss analysis of the vibrational structure for bis-
CF3-DPH. The resulting fits are given as solid lines; the individual
Gauss curves are given as dashed lines.
exchange hybrid functional48 and the correlation functional of
Lee, Yang, and Parr.49 Time-dependent density functional theory
1
(TDDFT)50 was employed for the Bu excited state geometry
optimization of the DPHs and the S1 state of Rh123. Unre-
stricted density functional theory (UDFT) was utilized for
determining the minimum structures in the corresponding T1
states. All geometry optimizations were performed with the
TURBOMOLE 5.7 program package.51,52 Solvent effects were
taken into account only in case of the cationic dyes. To this
end, explicit solvent molecules and a conductor-like screening
model (COSMO) were employed.53 The combined density
functional theory/multireference configuration interaction (DFT/
MRCI) method by Grimme and Waletzke54 was used for single-
point energy calculations at the obtained (TD)DFT equilibrium
geometries. Recently, this method was shown to yield the correct
order of excited states in R,ω-diphenylpolyenes55 and ꢁ-car-
otenes.56
The technical parameters of the MRCI calculations were as
follows. The 1s shells of all non-hydrogen atoms were kept
frozen in the electron correlation treatments. The MRCI
reference space was determined iteratively. The initial set was
spanned by all single and double excitations from the five
highest occupied molecular orbitals (HOMOs) to the five lowest
unoccupied orbitals (LUMOs) of the ground state Kohn-Sham
determinant. The final reference space consisted of all configu-
rations that contributed with a squared coefficient of at least
0.003 to one of the six lowest roots in either the Ag or Bu
irreducible representations (C2h-symmetric molecules), the six
lowest roots in either the Ag or Au irreducible representations
(Ci-symmetric molecules), or the 12 lowest roots (asymmetric
molecules) in the first DFT/MRCI run.
Figure 2. (A) Structures of DPH derivatives and (B) the Rh123
structure36 with six ethanol molecules.
product to precipitate, which was again solved in a little acetone
and precipitated once more by adding hexane. Yield: 180 mg
(7%). The absorption and fluorescence spectrum including a
multi Gauss analysis are shown in Figure 3.
Rh123 was purchased from Sigma Aldrich (R8004) and used
as received. Solutions of DPHs in ethanol were prepared freshly
and kept in the dark to avoid the photoinduced isomerization
to a cis-isomere of the polyene.
2.2. Calculations. For the DPHs the valence triple-ꢀ basis
set with (d) polarization function on all non-hydrogen atoms
and (p) polarization function on hydrogen (TZVP)46 was used.
Test calculations on Rh123 showed that excitation energies
changed only marginally when the atomic orbital basis set was
reduced to the less demanding split valence basis set with (d)
polarization function on all non-hydrogen atoms (SV(P)).46 For
reasons of computer efficiency all calculations on Rh123 were
therefore carried out in the smaller SV(P) basis. The ground
state geometries were determined using Kohn-Sham density
functional theory (DFT)47 with Becke’s three-parameter (B3LYP)
The suitability of three different solvent models for describing
the interaction between the Rh123 chromophore and the solvent
surrounding was tested at the ground-state geometry taking water
as an exemplary solvent: (I) Microsolvation with six explicit
water molecules that mimic the hydrogen bonds, (II) COSMO
that accounts for the solvent polarity, and (III) a combination
of I and II. It turned out that the red shifts brought about by
COSMO alone (0.06 eV for S1 and 0.08 eV for T1) are too
small, whereas model I yields red shifts (0.15 eV for S1 and
0.19 eV for T1) that are close to the solvent shifts (0.19 eV for
S1 and 0.23 eV for T1) obtained with our best model III.
Moreover, model I has the advantage that it is applicable in
excited state optimizations, too. We therefore chose model I
for the calculation of the adiabatic singlet and triplet excitation
energies of Rh123 in ethanol solution (Figure 2B).
2.3. Cyclic Voltammetry. Redox potentials were determined
by cyclic voltammetry (CV) in N,N-dimethylformamide (DMF).
DMF (water-free, under nitrogen) was purchased from Aldrich.
Tetra-n-butylammonium hexafluorophosphate (TBA PF6) was

4102 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Pfiffi et al.
used as received from Fluka and stored under argon. Dried argon
was bubbled through the test solution for 15 min prior to and
passed over the solution during the measurements. CV was
performed with a potentiostat (Metrohm, AUTOLAB PGSTAT
20) and special software (GPES, Eco Chemie, version 4.9). The
scan speed was 100 mV/s. A three electrode arrangement in a
single cell was used for measurements: a thin Pt-sheet as the
auxiliary electrode, a Ag/AgCl electrode as the reference
electrode (Metrohm, 6.0726.100, 11040486), and a glassy carbon
electrode (Metrohm, 6.1204.110) as the working electrode,
which was polished with Al2O3 powder in water after each
measurement. The reference electrode was filled with 100 mM
TBA PF6 in acetonitrile. Ferrocene had an oxidation potential
of 0.43 V against this reference electrode. As ferrocene has a
half-wave potential of 0.69 V versus the normal hydrogen
electrode (NHE), all reported potentials are cited against the
NHE by adding 0.26 V to the measured ones. In electrochemical
measurements, typical concentrations of the DPHs were
0.1-0.3 mM.
2.4. Steady-State Spectroscopy. All experiments were car-
ried out at 25 °C in spectroscopic grade THF (nondegassed).
Absorption spectra were recorded with a Varian-Cary 300
spectrometer. Fluorescence spectra were monitored with a
Fluorolog-3 spectrometer (Jobin Yvon). To exclude polarization
effects, the fluorescence of the probe was observed in a
conventional 90° setup with a polarizer set to the magic angle
(54.7°). For the measurements of the fluorescence quantum
yields, the spectra were corrected for wavelength-dependent
spectral sensitivity.
2.5. Fluorescence Correlation Spectroscopy. FCS was
carried out with a confocal epi-illuminated microscope similar
to that described before.15 The fluorescent molecules are excited
by a linearly polarized argon-ion-laser (Melles-Griot 35LAP431-
230) at 496 nm in cw mode. The laser is focused into the sample
by a water-immersion objective lens (UPLAPO 60 NA ) 1.2,
Olympus, Hamburg, Germany). In all measurements, the focal
plane was 50 µm away from the coverslip. The fluorescence is
collected by the same objective, separated from the excitation
by a polychroic beam splitter (498 DCLP, AHF, Tu¨bingen,
Germany), and is detected by two avalanche photodiodes
(Micro-Photon-Devices, PDM 50CT; Picoquant Berlin, Ger-
many) in a beam splitting arrangement to eliminate dead time
and afterpulsing artifacts. Fluorescence band-pass filters (HQ
533/46, AHF) block residual light and reduce Raman scattering
from the solvent. A confocal pinhole of 80 µm diameter yields
a characteristic diffusion time of 140 µs for Rh123 in water.
The signals of the two detectors are processed by a hardware
correlator (ALV-5000, ALV-Laser, Langen, Germany). The
measured correlation curves are fitted to different expressions
for the FCS curves as stated, using a Levenberg-Marquardt
nonlinear least-squares algorithm.
state population. The time-dependent parts of the normalized
correlation function can then be expressed as60
1/2
1
1
1
G(tc) ) 1 +
×
2
(
)
(
NF 1 + tc/td
)
1 + (ω0/z0) tc/td
(1 - Teq + Teq · e-t /t - Req + Req · e-t /t
) (2)
c
T
c R
NF is the mean number of fluorescent molecules within the
sample volume element. The molecule detection function (MDF)
is approximated by a three-dimensional Gauss distribution of
the detected fluorescence with radial and axial 1/e radii of ω0
and z0, respectively. The characteristic diffusion time td for the
fluorescent molecules is related to the translational diffusion
2
coefficient D by D ) ω0 /4td. Two bunching terms are possible:
(I) the triplet bunching term is always needed, where Teq is the
mean fraction of fluorophores within the sample volume element
being in their triplet states and tT is the relaxation time of the
triplet state, and (II) in case of radical formation a second
bunching term is needed,8 where Req is the mean fraction of
fluorophore radicals and 1/tR is the radical relaxation rate given
by the sum of rates for formation and decay of the radical. At
low irradiances typical parameters of the MDF for Rh123 in
ethanol are as follows for the used O2-N2 gas mixture: z0/ω0
) 5.3 and td ) 240 µs. These parameters were left free in the
fits of the data sets for a power series to compensate increasing
optical saturation, which distorts the MDF.15,61 As expected, both
parameters increase significantly with rising irradiance and
depend on the solvent as well as the degree of deoxygenation.
In the presence of micromolar concentrations of triplet
quenchers [Q] and making the simplifying assumption of a
uniform excitation profile within the detection volume, the
expressions for Teq and tT are given by15,60,62
k
ISCk01
Teq
)
(3)
(4)
(kISC + kT′)k01 + kT′k0
k01 + k0
(kISC + kT′)k01 + kT′k0
tT )
Considering a photophysical model with the three states S0,
S1, and T1, k0 is the deactivation rate of the first excited singlet
state S1 to the ground singlet state S0. kISC refers to the rate of
intersystem crossing from S1 to the lowest triplet state T1 and
kT′ is the effective rate of triplet state deactivation. The average
excitation rate within the detection volume element depends on
the average excitation intensity within the detection volume I0/2
and is given by k01 ) σ01(λ)γI0/2, where σ01 is the excitation
cross section, λ is the excitation wavelength, and γ ) λ/(hc) is
the reciprocal photon energy, with h being Planck’s constant
and c being the velocity of light. The average intensity I0/2 is
All measurements were performed in ethanol (Uvasol, Merck)
at room temperature (20 ( 1 °C). To adjust the triplet lifetime,
the solutions were purged with an O2-N2 gas mixture with an
oxygen content of approximately 0.1%.
2
determined by I0/2 ) P/(πω0 ), where P is the laser intensity
measured with a power meter.
The triplet quenchers with the bimolecular quenching constant
kqT influence the triplet deactivation by
3. Theory of Fluorescence Correlation Spectroscopy
FCS analyzes spontaneous fluorescence intensity fluctuations
of fluorescent molecules excited in a focused beam with the
correlation time tc.57-59 This fluctuation may be caused by a
broad range of dynamic processes at the molecular level:
changes in concentration of a fluorophore due to translation
diffusion of the fluorescent molecules into and out of the sample
volume element and changes in the excited singlet- or triplet-
kT′ ) kT + kqT[Q]
(5)
where kT is the effective rate of triplet state deactivation without
quencher. The fluorescence count rate per molecule Fcpm is
computed by normalizing the measured fluorescence signal F

DPH as Photoprotective Agent for Fluorescence Detection
TABLE 1: Computed Energies of Various DPHs in the Gas Phase in Electron Volts
vertical excitation at S0 minimuma adiabatic excitationb
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4103
vertical excitation at S1 minimumc,d
S1
S2
T1
T2
S1
T1
S2
S1
Bu/Au
Ag
Bu/Au
Ag
Bu/Au
Bu/Au
Ag
Bu/Au
DPH
3.34
3.28
3.01
3.36
3.05
3.29
3.18
3.51
3.46
3.28
3.52
3.26
3.46
3.43
1.81
1.77
1.71
1.81
1.69
1.77
1.75
2.85
2.80
2.63
2.85
2.64
2.78
2.79
3.10
3.06
2.81
3.13
2.84
3.06
2.97
1.39
1.37
1.33
1.40
1.31
1.37
1.36
2.79
2.75
2.69
2.81
2.60
2.75
2.80
2.94
2.90
2.68
2.97
2.69
2.92
2.84
CF3-DPH
Me2N-DPH
F-DPH
CN-DPH
bis-CF3-DPH
HMe2N+-DPH
a Comparable to the experimental absorption maximum (see Figure S5 and Table S1, SI). b Comparable to the 0-0 transition obtained from
Gauss analysis of the experimental data (see Figure S5 and Table S1, SI). c Comparable to the experimental fluorescence maximum (see Figure
S5 and Table S1, SI). d Except for Me2N-DPH, the order of the S1 and S2 states is reversed at this geometry.
TABLE 2: Experimental Energies of the Spectra of Various DPHs Measured in THF
absorption
fluorescence
a
c
ꢀ [M-1 cm-1
]
f (L)b
∆Emax [eV]c (λmax [nm])
E0,0 [eV]d
∆Emax [eV]c (λmax [nm])
ΦF
DPH
78 200
67 600
69 300
59 200
42 800
62 300
2.18
2.20
2.49
2.11
2.50
2.08
3.48(356)
3.44(361)
3.06(408)
3.50(354)
3.28(378)
3.44(360)
3.30
3.26
2.94
3.32
3.09
3.26
2.73(455)
2.86(434)
2.55(487)
2.74(452)
2.74(453)
2.86(433)
0.64
0.11
0.33
0.58
0.29
0.08
CF3-DPH
Me2N-DPH
F-DPH
CN-DPH
bis-CF3-DPH
a Extinction coefficients in ethanol. b Oscillator strengths obtained from DFT-MRCI calculations. c Maximum of the absorption/fluorescence
spectra and fluorescence quantum yields in THF. The quantum yield of DPH was taken from literature and used as reference standard for the
determination of the fluorescence quantum yields64 d Energy of the singlet state taken from the 0,0-transition from the multi-Gauss analysis (see
Figure S5 and Table S1, SI).
to the total number of molecules N ) NF/(1- Req - Teq) by
taking the triplet fraction Teq and the radical fraction Req into
account:
progression becomes smaller for higher excited states. The
UV-vis absorption and fluorescence spectra show only a minute
Stokes shift of approximately 600 ( 200 cm-1. All experimental
spectral properties are summarized in Table 2 (for a detailed
discussion, see sections 2.2 and 2.3 of the SI).
F
NF
Fcpm
)
(1 - Req - Teq)
(6)
The calculated vertical absorption energies are in good
agreement with the maximum of the absorption spectrum. It
should be noted, however, that the good agreement between
theory and experiment is somewhat fortuitous. Benchmark
calculations on linear polyenes and R,ω-diphenyl-polyenes55
have shown that our theoretical approach systematically under-
estimates the electronic excitation energies of these compounds
in the gas phase. This error is partially compensated by solvent
effects, which are known to be strong for the 11Bur11Ag
transition in DPH.65
Substitution in para or meta positions of the phenyl rings by
CF3 groups (CF3-DPH, bis-CF3-DPH) is found to have only
minor effects on the electronic spectrum, but changes the redox
potential significantly (see Table 3 below). In contrast,
dimethylamination of DPH in para position (Me2N-DPH) or
introduction of a cyano group (CN-DPH) in this position has a
large influence on the spectral properties. Both types of
substitution lead to an effective increase of the conjugation
lengths and a red shift of the absorption maximum. The obtained
data (Table 2, Figure S5 in the SI) show that the trends observed
for the vertical and adiabatic absorption energies of the
compounds are reproduced well by our theoretical approach.
On the average, the calculated vertical absorption energies are
too small by 0.1 to 0.15 eV, only for CN-DPH is the error
0.2 eV. For the adiabatic transitions, errors are slightly larger.
Singlet States of Rh123. Considering the fluorophore Rh123
in vacuum, vertical excitation energies for absorption
ES,vac(1F1r1F0) ) 2.70 eV and fluorescence ES,vac(1F1f1F0) )
2.60 eV were computed. However, since rhodamines are highly
4. Results and Discussion
4.1. Energies and Spectra. Singlet States of DPHs. Com-
puted vertical excitation energies of all DPHs at their respective
ground state geometries are collected in Table 1.
According to the calculations, the first excited singlet state
is the 11Bu state (1Au in the lower symmetric Me2N-DPH) in
the vertical absorption spectra of most compounds. It is mainly
characterized by a single excitation from the HOMO to the
LUMO. Geometry relaxation in the excited singlet state leads
to an equalization of C-C bond lengths in the chain, similar to
the situation in all-E-hexatriene.44 For details, see Figure S4 of
the SI.
The transition from the electronic ground state exhibits large
oscillator strength f (2.1-2.5, see Table 2). The computed
vertical excitation energy should be comparable to the
Franck-Condon maximum of the corresponding experimental
absorption spectrum.
The absorption and fluorescence spectra for all DPHs except
Me2N-DPH are very similar (see Figures 3 and S5). Multi-
Gaussian fits (full lines, Figures 3 and S5) were performed to
characterize the spectral features of all spectra and the corre-
sponding Stokes shifts. The progression of the lower vibronically
excited states is about 1400 cm-1 for all DPHs (see Table S1,
SI). This value is typical for polyenes and averages two different
vibrational modes at 1600 cm-1 (CdC stretch) and 1200 cm-1
(C-C stretch), which cannot be resolved in solution.63 The

4104 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Pfiffi et al.
TABLE 3: Measured Oxidation and Reduction Potentials and the Estimated Standard Free Energy of the Reductive Electron
Transfer by Eq 1 between Quencher and the S1, T1 and Radical Cation State R•+ of Rh123 in Ethanola
[V] vs NHE
∆GET [eV] reduction of Rh123
R•+
kqT [109 M-1 s-1
]
b
b
quencher
Eox
Ered
S1
T1
Me2N-DPH
F-DPH
DPH
CF3-DPH
CN-DPH
bis-CF3-DPH
0.57#
1.26
1.28
1.42
1.42
1.52
-1.92
-1.73
-1.71
-1.50
-1.29#
-1.44
-1.23
-0.54
-0.52
-0.38
-0.38
-0.28
-0.72
-0.03
-0.01
0.13
0.13
0.23
-0.64
0.05
0.07
0.20
0.20
0.30
6 ( 1c
2.9 ( 0.3
4.9 ( 0.6
2.5 ( 0.2
n.d.
2.1 ( 0.7
a All oxidative processes were estimated to be ∆GET > 0. For the calculation of the redox potentials of Rh123, Eox ) 1.21 V vs NHE and
Ered ) -0.61 V vs NHE, and the energies of the S1 and the T1 state, Es ) 2.41 eV and ET ) 1.90 eV were used. The contribution of the
Coulomb term is assumed to be negligible. The estimated errors for the measured potentials are <0.05 V. The precision for the quenching rates
corresponds to the statistical error of the linear fit for the slope (eq 5). The accuracy (systematic error) is mainly due to errors in the
determination of the quencher concentration via UV/vis absorption spectroscopy using the extinction coefficients in Table 2 (<5%). b All peaks
were irreversible unless they are indicated with a number sign (#). c Estimated quenching rate due to additional dark state; see section 4.3.
polar molecules, a distinct solvent effect is expected. Complex-
ation by six explicit ethanol molecules (Figure 2B, solvation
model I) results in a reduction of the energies to ES,solv(1F1f1F0)
) 2.44 eV (508 nm) and ES,solv(1F1f1F0) ) 2.32 eV (534 nm),
respectively. These calculated transition energies of Rh123 in
ethanol agree nicely with the experimental energies for the
maximum absorption and fluorescence, ∆EA ) 2.43 eV
(511 nm) and ∆EF ) 2.34 eV (530 nm), respectively.
Triplet States of DPHs. In the DPH derivates two triplet
states T1(Bu/Au) and T2(Ag) are found below S1 (see Table 1).
As for S1, the electronic structure of T1 is mainly characterized
by a single excitation from the HOMO to the LUMO. However,
unlike S1, geometry relaxation in the first excited triplet state
leads to an inversion of single and double bonds in the chain,
as observed previously for all-E-hexatriene55 (for details see SI
Section 2.1 and Figure S4). Within the accuracy of our
computational method, the vertical T1 excitation energy
ET(3Q1r1Q0) ) 1.76 ( 0.05 eV is the same for all DPHs. This
applies also to the adiabatic energy ET(3Q1f1Q0) ) 1.36 (
0.04 eV. Experimental values for the triplet energies ET of the
parent compound DPH are available from energy transfer
experiments in solution (ET,solv ) 1.47-1.54 eV).66-68 As the
computed energies agree satisfactorily with experimental values,
we assume that the accuracy of these triplet calculations is as
good as the ones for absorption and fluorescence of the singlet
states. It is noteworthy that the effect of substituents on the
adiabatic excitation energy of the T1 state is much less
pronounced than in the singlet case.
our DFT/MRCI results slightly underestimate the true excitation
energies but are in the right ballpark, in line with general trends
observed for this method.71
Free Energies for Energy Transfer Reactions. In order to
avoid singlet quenching of a fluorophore by SEET (see Figure
1B), the condition ES(F) < ES(Q) must be fulfilled for the singlet
energies. This condition is met even for the quencher Me2N-
DPH with the most red-shifted absorption maximum, for which
ES(1Q1r1Q0) was measured to be 3.01 eV, i.e. SEET of Rh123
to any of these DPHs is endergonic. For our DPHs, the optically
dark 21Ag state, which constitutes the first excited singlet state
in longer carotenes, is located slightly above the 11Bu/11Au state
at the ground-state geometry.
The condition for efficient triplet quenching of Rh123 by
TEET (see Figure 1B), ET(F) > ET(Q), is fulfilled by all DPHs,
and the spectral overlap between donor and acceptor should be
good.
4.2. Redox Properties. To judge whether photoinduced
electron transfer reactions can be a relevant quenching pathway,
the oxidation and reduction potentials, Eox and Ered, of the DPHs
were measured by CV in anhydrous DMF. The obtained
potentials are compiled in Table 3. Typical cyclic voltammo-
grams of all DPHs are shown in Figure S6 (SI). Moreover, the
standard free energy changes of the electron transfer reaction
∆GET of a DPH quencher with Rh123 in its S1, T1, or radical
cation state R•+ are given to check its energetic feasibility. The
reaction with R•+ also has to be considered, because it can be
formed by two-step photolysis at very high irradiances.15 The
species-specific∆GET valueswerecalculatedbytheRehm-Weller
eq 1 using the redox potentials and the excited state energies
of the involved states of Rh123.31
Even though most redox reactions are electrochemically
irreversible, the good linear correlation of the peak potentials
with the Hammett σ coefficients72 for the substituents of the
aromatic DPH ring system support the interpretation that these
potentials indeed reflect one-electron oxidation and reduction
properties (Figure 4). Both the oxidation and the reduction
potentials are reduced by electron donating groups (Me2N-DPH)
and increased by electron withdrawing groups (CF3-DPH, CN-
DPH, bis-CF3-DPH). The dotted line in Figure 4 indicates the
substituent-specific feasibility for a triplet state reduction of
Rh123. Quenchers with oxidation potentials that drop below
this limit are able to quench the triplet state of Rh123 by electron
transfer.
Triplet State of Rh123. The vertical triplet excitation energy
ET(3F1r1F0) of Rh123 at the S0-geometry was computed to
2.14 eV in vacuum and to 1.86 eV in ethanol. Similar to the
singlet case, the formation of hydrogen bonds causes a
substantial red shift of the excitation wavelength. At the triplet
minimum geometry, a vertical emission energy of 2.02 eV
(vacuum) and 1.76 eV (ethanol) is obtained. The calculated
3
adiabatic F1 emission energy in ethanol solution (solvation
model I) amounts to 1.82 eV. No experimental values have been
reported for the triplet energy of Rh123. However, experimental
triplet data (maximum of the phosphorescence spectrum,
∆EP ) 1.90 eV (651 nm)69,70) are available for the related
derivate Rhodamine 110 (Rh110), where the carboxyphenyl
residue is not esterified. We expect very similar molecular
energies because the maxima of the absorption and fluorescence
spectra of Rh123 are only slightly red-shifted by 1 to 2 nm
with respect to Rh110 (maxima of singlet absorption,
∆EA ) 2.43 eV (510 nm); fluorescence, ∆EF ) 2.35 eV
(528 nm)). Comparison of the experimental values for Rh110
with our computed excitation energies for Rh123 shows that
All oxidizing processes of the excited states S1, T1, and the
photo-oxidized state R•+ of Rh123 are estimated to be ender-
gonic (∆GET > 0). Therefore, only the free energies of the
reduction reactions of those states are given in Table 3. By

DPH as Photoprotective Agent for Fluorescence Detection
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4105
an important fluorescence quenching pathway that leads to the
formation of reactive singlet oxygen.
To discuss the distinct quenching effects on the T1 and S1
state, it is useful to define a quantum yield for quenching of
the individual state X by ΦqX ) (kqX [Q])/(kX + kqX [Q]). Here
kX is the inherent rate constant for the state deactivation, and
kqX is the bimolecular quenching constant. Considering a
quencher with diffusion controlled kdiff at a concentration [Q]
) 10 µM and the distinct S1 and T1 lifetimes of Rh123 with
1/k0 ) 4 ns and 1/kT ) 40 µs, the quantum yield of singlet
quenching ΦqS ) 0.02% is more than 3000 times smaller than
the quantum yield of triplet quenching ΦqT ) 68%. This
selective quenching effect for DPH leads to a concomitant
increase of the fluorescence count rate per molecule Fcpm (more
than three times) and decrease of Teq with increasing DPH
concentration (Figure 5B and filled squares in Figure 6).
Other DPHs. In the following, we tested several substituted
DPHs by FCS (CF3-DPH Figure S7, F-DPH Figure S8, bis-
CF3-DPH Figure S9). The obtained triplet quenching constants
kqT are also listed in Table 3. Even if the limited accuracy
(systematic error) may give errors of up to 5% and the precision
(shot noise) is on average (0.3 × 109 M-1 s-1, the difference
of the quenching constants (approximately a factor of 2) is still
significant and shows a systematic trend: The quenching constant
decreases with increasing size or electronegativity of the DPHs.
Moreover, considering size effects, the Smoluchowski equation
predicts the opposite effect: the bimolecular quenching constant
should rise and not decrease with increasing radius of the
quencher; i.e., for the molecular dimensions of Rh123 and the
different DPHs (maximal radius increase 16%) an increase of
kqT of maximal 4.4% could be expected.
Figure 4. Oxidation and reduction potentials measured in DMF were
plotted versus the Hammett coefficients. The linear relationship between
the potentials and the Hammett coefficients is given by Eox ) 1.13 +
0.52 × σ and Ered ) -1.68 + 0.35 × σ. The dotted line indicates the
feasibility of the triplet state reduction of Rh123. Quenchers with
oxidation potentials that drop below this limit are able to quench the
triplet state of Rh123 by electron transfer. In order to complete the
Hammett plot on the side of the electron donating groups an additional
methoxy-substituted DPH (MeO-DPH) (Eox ) 0.99 V vs NHE, Ered
) -1.83 V) with a σ value of -0.27 was measured. For further details,
see SI.
introducing more electron withdrawing groups the quenching
of the reduction reactions becomes less favorable. In order to
avoid reduction of the S1 state of Rh123 completely, even more
electron deficient DPHs would be necessary. The T1 state of
Rh123 is expected to be also quenched effectively by electron
transfer especially by Me2N-DPH (dotted line for ∆GET ) 0 in
Figure 4), whereas bis-CF3-DPH should not quench via electron
transfer but still via TEET (for more discussions, see section
4.3).
4.3. Fluorescence Correlation Spectroscopy Measures
Triplet Quenching. As the solubility of DPHs in water is very
low, we studied them in ethanol. Because of the high solubility
of the triplet quencher oxygen in ethanol, the rate constant for
triplet relaxation kT of Rh123 is large, which results in a very
low stationary triplet population in air-saturated solution.
Therefore we bubbled the sample solution with a premixed
O2-N2 gas mixture, so that the oxygen concentration is reduced
from 20% (air) to approximately 0.1%. To determine triplet
parameters by FCS using eqs 2-4, a series of correlation curves
at different mean irradiances I0/2 was measured.
Another important aspect for applications as photoprotective
agents is the solubility of triplet quenchers. In the case of bis-
CF3-DPH, we observe a saturation of the beneficial effect above
10 µM. Carotenoids and carotene derivatives are known to form
complex aggregates in polar solvents already at low concentra-
tions.76 Therefore, in further studies, substituents should be used
that improve the solubility of the DPHs in polar solvents or
water.
Effects of the Electron-Rich DPH: Me2N-DPH. Figure 7A
shows the FCS curves for Rh123 in ethanol at a quencher
concentration of 8 µM Me2N-DPH. The correlation curves G(tc)
of Rh123 measured in the presence of the quencher could not
be described by eq 2 with a single bunching term. An additional
bunching term is needed at a relaxation time of tR ) 77 µs with
an amplitude Req ) 0.30. Even though the decreased triplet
amplitude Teq indicates efficient triplet quenching (Figure 7B,
upper panel, kqT ) 6 ( 1 × 109 M-1 s-1), surprisingly a decrease
in fluorescence signal Fcpm is observed (Figure 7B, lower panel).
Considering the possible electron transfer pathways in Table
3,one expects the formation of Rh123 free radical anions R•-
resulting from the reduction of the triplet state of the fluorophore
by Me2N-DPH. As the oxygen concentration is low, this process
produces another relatively long-lived dark state and competes
with the beneficial effect of triplet quenching.
DPH. Varying the irradiance, a typical set of correlation
curves is displayed in Figure 5A for Rh123 in the presence of
the quencher DPH at a concentration of 15.3 µM.
The analysis of correlation curves by eq 2 shows by equally
distributed residuals that a diffusion term with the time td and
a single triplet bunching term with the relaxation time tT are
sufficient to describe the data at low and medium irradiances.
DPH is indeed a triplet quencher of Rh123, because the triplet
amplitude Teq (Figure 5C) and triplet time tT decrease with
increasing DPH concentration (Figure 5D). Global analysis of
Teq and tT by eq 3 and 4 allowed us to compute kISC and kT′ for
different DPH concentrations. Figure 5E shows that kT′ depends
linearly on the DPH concentration as expected by eq 5, whereas
kISC is independent of the quencher concentration [Q]. The
determination of the slope yields a triplet quenching constant
kqT of DPH ) 4.9 × 109 M-1 s-1, which is close to the diffusion
controlled value kdiff ) 5.4 × 109 M-1 s-1 66. It is important to
note that kISC(0.1% O2) ) 2 × 105 s-1 is 4.5 times smaller than
kISC(air). In agreement with previous findings,14,73-75 this
indicates that, for Rh123 oxygen enhanced ISC, there is also
Next we changed the electron-donating dimethylamino groups
in Me2N-DPH to electron-withdrawing quaternary ammonium
groups to decrease the electron density of the quencher and thus
to counteract its ability to reduce the T1 state of Rh123. At the
same time, the solubility should be significantly increased. Using
the Hammett σ-coefficient for NH3+ (σ ) 0.86) and the relation
of Figure 4, we are able to estimate Eox ) 1.58 V vs NHE,

4106 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Pfiffi et al.
Figure 5. (A) Normalized correlation curves of Rh123 in ethanol ([O2] ∼ 0.1%) with a quencher concentration of [Q] ) 15.3 µM at different laser
irradiances I0/2. (B) Count rate per molecule Fcpm. (C,D) The parameters of the average molecule fraction in the triplet state Teq and the triplet
correlation time tT according to eq 2 at 0, 4.85, 11.5, and 15.3 µM DPH were plotted against the applied laser irradiance. The triplet parameters
were fitted globally to eqs 3 and 4 to give kISC and kT′ for each concentration. (E) The slope of the observed triplet relaxation rates kT′ with the
concentration is equal to the triplet quenching rate kqT ) 4.9 × 109 M-1 s-1
.
Two derivatives of DPH with quaternary ammonium groups
have been studied (see Section 1, SI): (I) Me2N-DPH was
dissolved in ethanol containing 1% sulfuric acid to give
HMe2N+-DPH ((all-E)-1,6-bis-[4-(dimethylammonium)phenyl-
]hexa-1,3,5-triene bis hydrogen sulfate) (Figure S2, the acidic
conditions do not affect the cationic dye Rh123). (II) An
alkylated DPH with a quaternary ammonium group was
synthesized ((all-E)-1,6-bis-[4-(ethyldimethylammonium)phe-
nyl]hexa-1,3,5-triene bis p-toluenesulfonate EtMe2N+-DPH)
(Figure S3). It is noteworthy to mention that the specific spectral
red shift in absorption of Me2N-DPH vanishes completely upon
quaternization of the amino groups (see Figures S2 and S3).
The resulting UV/vis absorption spectra are now identical to
the other DPHs. Considering HMe2N+-DPH, the computed
vertical T1 energy for absorption ET(3Q1r1Q0) ) 1.75 eV does
not deviate from the other DPHs (see section 4.1), so that TEET
should be possible. Surprisingly, FCS experiments with HMe2N+-
DPH and EtMe2N+-DPH showed no changes in signal strength
or triplet population of Rh123. No differences were found in
the FCS curves of Rh123 measured in ethanol or ethanol with
sulfuric acid as control. The count rate per molecule as well as
the triplet amplitude and the diffusion time remained unchanged.
Figure 6. Dependence of the fluorescence signal Fcpm and the triplet
population Teq of Rh123 on the concentration of DPH (9) and bis-
CF3-DPH (0) at a laser irradiance of I0/2 ) 100 kW/cm2.
which is similar to the potential of bis-CF3-DPH (see Table
3). Thus, no triplet quenching by electron transfer should be
possible.

DPH as Photoprotective Agent for Fluorescence Detection
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4107
bis-CF3-DPH. Moreover, the Coulomb repulsion between the
two positively charged molecules RMe2N+-DPH and Rh123
increases their interaction distance, which explains the complete
loss of the ability for triplet quenching. To conclude, the ability
of DPH derivatives to quench dye triplet states is governed by
a delicate interplay of steric, Coulombic and electronic effects.
Electron deficient DPHs have indeed beneficial effects on the
fluorescence rate of Rh123 by reducing the triplet state
population and by avoiding the formation of other long-lived
dark states.
Acknowledgment. The present work has been performed as
a project of the SFB 663 (A8, C1) at the Heinrich-Heine-
Universita¨t Du¨sseldorf. We thank Jerker Widengren and Ralf
Ku¨hnemuth for very helpful discussions on FCS. This work is
dedicated to the memory of Prof. Dr. Hans-Dieter Martin, who
passed away on March 8, 2009.
Supporting Information Available: Section 1 contains a
detailed description of the synthesis and characterization for the
different DPHs. In Section 2, further information on excited
state relaxation of the DPHs (2.1), the absorption (2.2), and
the emission spectra (2.3) of the DPHs as well as the multi
Gaussian analysis of these (2.2 and 2.3) are given. Furthermore,
Section 2 gives the cyclic voltammograms for the DPHs (2.4)
and the FCS data for the DPHs not shown in the main
manuscript (2.5). This material is available free of charge via
the Internet at http://pubs.acs.org.
Figure 7. (Α) Normalized correlation curves of Rh123 in ethanol at
I0/2 ) 200 kW/cm2. In the presence of 8 µM Me2N-DPH, an additional
bunching term is needed to describe the curve. The fitted relaxation
parameters were Teq ) 0.76 and tT ) 11 µs for Rh123 and Teq ) 0.53,
tT ) 7.7 µs, Req ) 0.30 and tR ) 77 µs for Me2N-DPH. (B) Average
fraction of molecules in the triplet state Teq and fluorescence count rate
per molecule Fcpm as a function of the applied laser irradiance.
References and Notes
(1) Yamashita, M.; Kuniyasu, A.; Kashiwagi, H. J. Chem. Phys. 1977,
66, 986.
(2) Drexhage, K. H.; Scha¨fer, F. P. Structure and properties of laser
dyes. In Topics in Applied Physics; Springer Verlag: Berlin, 1973; Vol. 1,
pp 144.
(3) Gonc¸alves, M. S. T. Chem. ReV. 2009, 109, 190.
(4) Pawley, J. B. Handbook of Biological Confocal Microscopy, 3rd
ed.; Springer: New York, 2006.
It seems reasonable that the positive charges of the two
RMe2N+-DPH derivatives are responsible for the lack of activity
(see Conclusions).
5. Conclusions
(5) Hell, S. W. Science 2007, 316, 1153.
(6) Widengren, J.; Kudryavtsev, V.; Antonik, M.; Berger, S.; Gerken,
M.; Seidel, C. A. M. Anal. Chem. 2006, 78, 2039.
(7) Haustein, E.; Schwille, P. Annu. ReV. Biophys. Biomol. Struct. 2007,
36, 151.
(8) Widengren, J.; Chmyrov, A.; Eggeling, C.; Lo¨fdahl, P. A.; Seidel,
C. A. M. J. Phys. Chem. A 2007, 111, 429.
(9) Vogelsang, J.; Cordes, T.; Forthmann, C.; Steinhauer, C.; Tinnefeld,
P. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 8107.
(10) Hofkens, J.; Cotlet, M.; Vosch, T.; Tinnefeld, P.; Weston, K. D.;
Ego, C.; Grimsdale, A.; Mu¨llen, K.; Beljonne, D.; Bredas, J. L.; Jordens,
S.; Schweitzer, G.; Sauer, M.; De Schryver, F. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 13146.
(11) Liphardt, B.; Lu¨ttke, W. Opt. Commun. 1983, 48, 129.
(12) Sauer, M.; Han, K. T.; Mu¨ller, R.; Schulz, A.; Tadday, R.; Seeger,
S.; Wolfrum, J.; Arden-Jacob, J.; Deltau, G.; Marx, N. J.; Drexhage, K. H.
J. Fluoresc. 1993, 3, 131.
(13) Ernsting, N. P.; Kaschke, M.; Kleinschmidt, J.; Drexhage, K. H.;
Huth, V. Chem. Phys. 1988, 122, 431.
(14) Eggeling, C.; Widengren, J.; Rigler, R.; Seidel, C. A. M. Photo-
stabilities of fluorescent dyes for single-molecule spectroscopy: Mechanisms
and experimental methods for estimating photobleaching in aqueous solution.
In Applied Fluorescence in Chemistry, Biology and Medicine; Springer:
Berlin, 1999; pp 193.
(15) Eggeling, C.; Widengren, J.; Rigler, R.; Seidel, C. A. M. Anal.
Chem. 1998, 70, 2651.
(16) Widengren, J.; Rigler, R. Bioimaging 1996, 4, 149.
(17) Vogelsang, J.; Kasper, R.; Steinhauer, C.; Person, B.; Heilemann,
M.; Sauer, M.; Tinnefeld, P. Angew. Chem., Int. Ed. 2008, 47, 5465.
(18) Rasnik, I.; McKinney, S. A.; Ha, T. Nat. Methods 2006, 3, 891.
(19) Ferguson, M. W.; Beaumont, P. C.; Jones, S. E.; Navaratnam, S.;
Parsons, B. J. Phys. Chem. Chem. Phys. 1999, 1, 261.
(20) Ringemann, C.; Scho¨nle, A.; Giske, A.; von Middendorff, C.; Hell,
S. W.; Eggeling, C. ChemPhysChem 2008, 9, 612.
(21) Fo¨rster, T. Discusssions Faraday Soc. 1959, 7.
(22) Fo¨rster, T. Ann. Phys. 1948, 2, 55.
DPHs quench the triplet state of Rh123 effectively with
specific quenching rates varying by more than a factor of 2 from
nearly diffusion-controlled (4.9 × 109 M-1 s-1) to lower values.
Possible reasons for the observed differences raise the question
of the detailed mechanism. The best quenching effect is obtained
with the parent compound DPH, whereas the quenching
efficiency is reduced with larger substituents and increasing
oxidation potentials. Two channels for triplet state deactivation
by the DPHs can be discussed:77,78 First, looking at the energetic
levels from the quantum mechanical calculations, the free energy
for Dexter TEET from the fluorescent dye to the DPHs ∆GTEET
should be exergonic (∆GTEET < -0.2 eV, see section 4.1 and
Table 1). Beyond that, triplet quenching by an additional electron
transfer pathway17 is considered to be feasible for those DPHs,
where the free energy ∆GET(T1) is exergonic (see Table 3). In
case of Me2N-DPH (∆GET(T1) ) -0.77 eV), a significant
population of a newly long-lived dark state is detected by FCS,
which could be assigned to free radicals. DPH and F-DPH differ
in kqT, but they have approximately the same ∆GTEET and
∆GET(T1). Thus additional effects by substituents, which are
important for the Dexter ET by influencing the spectral overlap,
must be considered as well: (I) Altering steric hindrance,79,80
(II) change in electron density in the relevant orbitals and orbital
overlap,81 or (III) introduction of intermolecular Coulombic
attraction or repulsion.
For example, larger steric hindrance by the CF3 groups might
contribute to the reduced quenching efficiency of CF3-DPH and

4108 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Pfiffi et al.
(23) Fo¨rster, T. Fluoreszenz Organischer Verbindungen; Vandenhoeck
and Ruprecht: Göttingen, Germany, 1951.
(53) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799.
(54) Grimme, S.; Waletzke, M. J. Chem. Phys. 1999, 111, 5645.
(55) Marian, C. M.; Gilka, N. J. Chem. Theory Comput. 2008, 4, 1501.
(56) Kleinschmidt, M.; Marian, C. M.; Waletzke, M.; Grimme, S.
J. Chem. Phys. 2009, 130, 044708.
(24) Dexter, D. L. J. Chem. Phys. 1953, 21, 836.
(25) Rehm, D.; Weller, A. Israel J.Chem. 1970, 8, 259.
(26) Rehm, D.; Weller, A. Z. Phys. Chem. 1970, 69, 183.
(27) Marcus, R. A. J. Chem. Phys. 1956, 24, 966.
(28) Kavarnos, G. J.; Turro, N. J. Chem. ReV. 1986, 86, 401.
(29) In order to describe the transitions involved in these processes we
followed IUPAC conventions and denoted the excited state first and the
ground state last. Thus absorption is marked by “r” and an emission is
marked by “f”.
(30) Mills, I.; Cvitas, T.; Homann, K.; Kallay, K.; Kuchitsu, K.
Quantities, Units and Symbols in Physical Chemistry, 2nd ed.; Blackwell
Science: Oxford, U.K., 1993.
(31) Wardman, P. J. Phys. Chem. Ref. Data 1989, 18, 1637.
(32) Pol´ıvka, T.; Sundstro¨m, V. Chem. ReV. 2004, 104, 2021.
(33) Furuichi, K.; Sashima, T.; Koyama, Y. Chem. Phys. Lett. 2002,
356, 547.
(34) Hilinski, E. F.; McGowan, W. M.; Sears, D. F.; Saltiel, J. J. Phys.
Chem. 1996, 100, 3308.
(35) Ye, J. F.; Chen, H.; Note, R.; Mizuseki, H.; Kawazoe, Y. Int. J.
Quantum Chem. 2007, 107, 2006.
(36) For the calculations regarding RH123, the ethyl ester was used
instead of the methyl ester.
(37) Sonoda, Y.; Kawanishi, Y.; Ikeda, T.; Goto, M.; Hayashi, S.;
Yoshida, Y.; Tanigaki, N.; Yaso, K. J. Phys. Chem. B 2003, 107, 3376.
(38) Zhuang, Z. P.; Kung, M. P.; Kung, H. F. J. Med. Chem. 2006, 49,
2841.
(39) Trepanier, V. E.; Fillion, E. Organometallics 2007, 26, 30.
(40) Kauffman, J. M.; Moyna, G. J. Org. Chem. 2003, 68, 839.
(41) Alford, P. C.; Palmer, T. F. Chem. Phys. Lett. 1982, 86, 248.
(42) Alford, P. C.; Palmer, T. F. Chem. Phys. Lett. 1986, 127, 19.
(43) Sonoda, Y.; Kwok, W. M.; Petrasek, Z.; Ostler, R.; Matousek, P.;
Towrie, M.; Parker, A. W.; Phillips, D. J. Chem. Soc. Perkin Trans. 2 2001,
308.
(57) Magde, D.; Elson, E. L.; Webb, W. W. Biopolymers 1974, 13, 29.
(58) Elson, E. L.; Magde, D. Biopolymers 1974, 13, 1.
(59) Magde, D.; Elson, E. L.; Webb, W. W. Phys. ReV. Lett. 1972, 29,
705.
¨
(60) Widengren, J.; Mets, U.; Rigler, R. J. Phys. Chem. 1995, 99, 13368.
(61) Enderlein, J.; Gregor, I.; Patra, D.; Dertinger, T.; Kaupp, U. B.
ChemPhysChem 2005, 6, 2324.
(62) Eggeling, C.; Volkmer, A.; Seidel, C. A. M. ChemPhysChem 2005,
6, 791.
(63) Christensen, R. L.; Barney, E. A.; Broene, R. D.; Galinato, M. G. I.;
Frank, H. A. Arch. Biochem. Biophys. 2004, 430, 30.
(64) Dupuy, B.; Montagu, M. Analyst 1997, 122, 783.
(65) Kohler, B. E.; Itoh, T. J. Phys. Chem. 1988, 92, 5120.
(66) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochem-
istry; Marcel Dekker Inc.: New York, 1993.
(67) Saltiel, J.; Crowder, J. M.; Wang, S. J. J. Am. Chem. Soc. 1999,
121, 895.
(68) Evans, D. F.; Tucker, J. N. J. Chem. Soc., Faraday Trans. 2 1972,
68, 174.
(69) Heupel, M.; Gregor, I.; Becker, S.; Thiel, E. Int. J. Photoenergy
1999, 1, 165.
(70) Menzel, R.; Bornemann, R.; Thiel, E. Phys. Chem. Chem. Phys.
1999, 1, 2435.
(71) Kunavin, N. I.; Nurmukhametov, R. N.; Khachaturova, G. T. Zh.
Prikl. Spektrosk. 1977, 26, 1023.
(72) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(73) Gregor, I.; Heupel, M.; Thiel, E. Chem. Phys. 2001, 272, 185.
(74) Stracke, F.; Heupel, M.; Thiel, E. J. Photochem. Photobiol. A 1999,
126, 51.
(75) Schweitzer, C.; Schmidt, R. Chem. ReV. 2003, 103, 1685.
(76) Ko¨psel, C.; Moltgen, H.; Schuch, H.; Auweter, H.; Kleinermanns,
K.; Martin, H. D.; Bettermann, H. J. Mol. Struct. 2005, 750, 109.
(77) Balzani, V.; Bolletta, F.; Scandola, F. J. Am. Chem. Soc. 1980,
102, 2152.
(78) Nau, W. M.; Scaiano, J. C. J. Phys. Chem. 1996, 100, 11360.
(79) Wamser, C. C.; Medary, R. T.; Kochevar, I. E.; Turro, N. J.; Chang,
P. L. J. Am. Chem. Soc. 1975, 97, 4864.
(80) Cheng, G.; Peng, X.; Hao, G.; Kennedy, V. O.; Ivanov, I. N.;
Knappenberger, K.; Hill, T. J.; Rodgers, M. A. J.; Kenney, M. E. J. Phys.
Chem. A 2003, 107, 3503.
(81) Albinsson, B.; Ma˚rtensson, J. J. Photochem. Photobiol. C 2008, 9,
138.
(44) Sonoda, Y.; Morii, H.; Sakuragi, M.; Suzuki, Y. Chem. Lett. 1998,
349.
(45) Barluenga, J.; Moriel, P.; Aznar, F.; Valdes, C. AdV. Synth. Catal.
2006, 348, 347.
(46) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829.
(47) Kohn, W.; Sham, L. J. Phys. ReV. 1965, 140, A1133.
(48) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(49) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(50) Furche, F.; Ahlrichs, R. J. Chem. Phys. 2002, 117, 7433.
(51) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Kolmel, C. Chem. Phys.
Lett. 1989, 162, 165.
(52) Schade, B.; Hagen, V.; Schmidt, R.; Herbrich, R.; Krause, E.;
Eckardt, T.; Bendig, J. J. Org. Chem. 1999, 64, 9109.
JP909033X